JP2002508609A - Method and system for processing directional sound in a virtual sound environment - Google Patents

Abstract

An acoustic virtual environment is processed in an electronic device. The acoustic virtual environment comprises at least one sound source ( 300 ). In order to model the manner in which the sound is directed, a direction dependent filtering arrangement ( 306, 307, 308, 309 ) is attached to the sound source, whereby the effect of the filtering arrangement on the sound depends on predetermined parameters. The directivity can depend on the frequency of the sound.

Description

DETAILED DESCRIPTION OF THE INVENTION

The present invention relates to a method and a system in which an artificial audible impression corresponding to a certain space can be generated for a listener. More particularly, the present invention relates to the processing of directional sounds in such hearing sensations and the resulting transmission of hearing sensations in systems where the information presented to the user is transmitted, processed and / or compressed in digital form. .

BACKGROUND ART [0002] A virtual acoustic environment refers to a sense of hearing that helps a listener to hear an electrically reproduced sound in a certain space. Complex virtual acoustic environments are often intended to mimic real space. It is called auditoryization of the space. This concept is described in the paper, M.E. Clynenel, Be. Eye. Darenbeck, PA. Svenson, "Hearing-An Overview", 1993, Yar. Audio Engineering Society, Vol. 41, N
o. 11, pp. 861-875 (M. Kleiner, BI Dalenbaeck, P. Svensson; "
Auralization-An Overview ”, 1993, J. Audio Eng. Soc., Vol. 41, No. 11,
pp. 861-875). Hearing can be combined in a natural way with the creation of a visual virtual environment, so that a user with a suitable display and speakers or a headset can observe the desired real or imaginary space and You can even "move around." Thus, the user obtains various visual and auditory sensations depending on the position in the environment selected as the observation point.

[0003] The creation of a virtual acoustic environment can be divided into three components: sound source modeling, spatial modeling, and listener modeling. The invention relates in particular to sound source modeling and early sound reflection.

The VRML97 language (Virtual Reality Modeling Language)
deling Language) 97) is commonly used to model and process visual and virtual acoustic environments, and this language is published in the ISO / IEC JTC / SC24 IS 14772-1, 1997, "Information Technology-Computer Graphics and Image Processing-Virtual Reality Modeling Language (Information Technology-Computer Graphics and Image Pro
cessing-The Virtual Reality Modeling Language (VRML97), April 1997 and Internet address http://www.vrml.org/Specifications/VRML97
It is handled on the corresponding page of /. Another set of rules being developed during the time this patent application was prepared is related to Java3D, which is a control and processing environment for VRML, for example, the publication SUN Publication 1997;
And Internet address http://www.javasoft.com/-products/java-me
dia / 3D / forDevelopers / 3Dguide /- Further, the developing MPEG-4 standard (Motion Picture Experts Group 4) states that multimedia presentations transmitted over digital communication links can include real and virtual objects,
They aim to form an audiovisual environment together. The MPEG-4 standard is published in the ISO / IEC JTC / SC29 WG11 CD 14496, 1997; "Information Technology-Coding of audiovisual objects." November 1997 and Internet address http: // www.cselt.it/-mpeg/p
It is described on the corresponding page of ublic / mpeg-4_cd.htm.

FIG. 1 shows a known directional acoustic model used in VRML97 and MPEG-4. The sound source is located at point 101, around which two ellipsoids 102 and 103 are assumed, one inside the other, so that the focus of one ellipsoid is in common with the position of the sound source and the two ellipsoids The main axes are parallel. Ellipsoids 102 and 104
The magnitude of the distance maxBack, maxFront, measured in the direction of the main axis,
Represented by minBack and minFront. The sound decay as a function of distance is represented by curve 104. The intensity of the sound is constant inside the inner ellipsoid 102, and is zero outside the outer ellipsoid 103. As one moves away from point 101 along all straight lines passing through point 101, the sound intensity decreases linearly by 20 dB between the inner and outer ellipsoids. In other words, the attenuation A observed at point 105 located between the two ellipsoids can be calculated by:

A = −20 dB · (d ′ / d ″) where d ′ is the distance from the surface of the inner ellipsoid measured along the straight line connecting the points 101 and 105 to the observation point, d "is the distance between the inner and outer ellipsoids measured along the same straight line.

In Java3D, a directional acoustic model is modeled by the conical acoustic concept shown in FIG. This figure represents a cross section of the two conical structures along a plane containing the common longitudinal axis of the cones. The sound source is located at the common vertex 203 of the cones 201 and 202. In both the front cone 201 and the rear cone 202, the sound is evenly attenuated. In the region between the two cones, a linear interpolation is applied. To calculate the attenuation detected at the observation point 204, the sound intensity without attenuation, the width of the front and rear cones, and the line between the longitudinal axis of the front cone and points 203 and 204 Need to know the angle.

[0008] A known method for modeling acoustic characteristics of a space having an acoustic reflecting surface is an imaginary sound source method (im
age source method), where a set of virtual imaginary sound sources, which are mirror images of the sound source corresponding to the reflection surface of the observation target, are provided in addition to the first sound source. Since one imaginary sound source is arranged after the reflection surface of each investigation object, the distance measured straight from the imaginary sound source to the observation point is the same as the distance from the first sound source to the observation point. Further, the sound from the imaginary sound source reaches the investigation point from the same direction as the actual reflected sound. The auditory sensation is obtained by adding the sound generated by the imaginary sound source.

The method according to the prior art is very computationally intensive. Assuming that the virtual environment is transmitted to the user, for example, over a broadcast or data network, the user's receiver will need to constantly add sound generated by thousands of virtual sources.
Moreover, if the user decides to change the position of the observation point, the basis of the calculation will always change. In addition, the known solution completely ignores the fact that, besides the direction angle, the directivity of the sound is strongly dependent on its wavelength, in other words, the fact that the frequency rises and falls the different sounds go in different directions. ing.

[0010] Finnish Patent Application 974006 (Nokia Corp.) describes a method and system for processing a virtual acoustic environment. There, the acoustic reflection surface of the environment to be modeled is represented by a filter having a constant frequency response. In order to transmit the modeled environment in digital transmission form, it is sufficient to express in some way the transfer functions of all the essential acoustic reflection surfaces belonging to the environment. However, even this does not consider the effect of the arrival direction or pitch of the sound on the direction of the sound.

It is an object of the present invention to show a method and system in which a virtual acoustic environment can be transmitted to a user with a reasonable computational load. It is a further object of the present invention to provide a method and system that can take into account the effect of pitch and direction of sound on the directivity of the sound.

The object of the present invention was characterized by parameters which set the desired directivity of the sound with the help of various parameters and considered how that directivity depends on the frequency and the direction of arrival. This is achieved by modeling the sound source or its early reflections with a system function.

[0013] The method according to the invention provides a method for modeling a direction in which sound is directed, such that the effect of the set of filters on the sound source depends on a predetermined parameter of the filter, so that the filter depends on the direction. Are provided corresponding to the sound sources of the virtual acoustic environment.

The present invention also relates to a system including means for generating a filter bank including a filter characterized by parameters for modeling directivity of sound from a sound source belonging to a virtual acoustic environment.

According to the invention, the model of the sound source or the reflection calculated therefrom comprises a direction-dependent digital filter. One reference direction, called the zero azimuth, is selected for the sound. This direction can be directed in any direction in the virtual acoustic environment. In addition, a number of other directions are selected, where it is necessary to model in which direction the sound is directed. Further, these directions can be arbitrarily selected. Each other direction selected is modeled by a unique digital filter with a transfer function that can be chosen to be frequency dependent or independent. If the observation point is located somewhere other than the direction just represented by the filter, it is possible to form various interpolations between the filter transfer functions.

When trying to model sound and how it is directed in a system where information must be transmitted in digital form, only the data for each transfer function need be transmitted. The receiving device knows the required observation point and determines with the help of the reconstructed transfer function that the sound is directed from the position of the sound source towards the observation point. When the position of the observation point changes with respect to the zero bearing, the receiver checks how the sound is directed to the new observation point. Since there may be several sound sources, the receiving device calculates how the sound is directed from each sound source to the observation point and modifies the reproduced sound accordingly. At that time, for example, for a virtual orchestra in which the musical instrument is located in various places and faces in various directions, the listener obtains a listening sensation in a correctly positioned listening position.

The simplest alternative to implementing direction-dependent digital filtering is to attach a gain to each of the selected directions. However, the pitch of the sound is not considered at that time. In a more refined alternative, the observed frequency bands are divided into sub-bands, each of which is given its own amplification factor in each direction selected for each sub-band. In a further refined version, each observed direction is modeled by a generalized transfer function, and a set of coefficients corresponding to the transfer function that allows reconstruction of the same transfer function is indicated.

In the following, the invention will be explained in more detail with reference to preferred embodiments, given by way of example, and to the drawings, in which:

In connection with the prior art, the illustrations of FIGS. 1-2 are made earlier, and in the following description of the invention, preferred illustrations are given in FIGS. 3-7b.

FIG. 3 shows the location of the sound source at the point 300 and the direction 301 of the zero azimuth. In the figure, it is assumed that the sound source located at point 300 is represented by four filters. The first filter represents sound propagating in the direction 302 from the source, the second filter represents sound propagating in the direction 303 from the source, the third filter represents sound propagating in the direction 304 from the source, and The filter at 4 represents sound propagating in the direction 305 from the sound source. Further, in the figure, it is assumed that the sound propagates symmetrically with respect to the zero azimuth direction 301. As a result, each of the directions 302-305 actually represents any corresponding direction on the conical surface obtained by rotating a radius representing the observed direction about the zero azimuth direction 301. The present invention is not limited to these assumptions, and some features of the present invention will be more readily understood by first considering a simplified embodiment. In the figure, directions 302 to 305 are shown as straight lines separated by the same distance in the same plane, but the directions can be arbitrarily selected.

Each filter representing sound propagating in a direction different from the zero azimuthal direction shown in FIG. 3 is symbolically indicated by blocks 306, 307, 308 and 309. Each filter is characterized by a transfer function H _i, where _i {1,2,3,4}. The transfer function of the filter is normalized such that the sound propagating in the zero azimuth is the same as the sound generated by the sound source as described above. Normally, sound is a function of time, so the sound generated by the sound source is denoted X (t). Each filter 30
6 to 309 generate a response Yi (t) (where i {1, 2, 3, 4}) according to the following equation.

Yi (t) = Hi ^* X (t) (1) Here, * represents convolution with respect to time. Response Yi (
t) is a sound directed in that direction.

In its simplest form, the transfer function means that the impulse X (t) is multiplied by a real number. Since it is natural to select the zero azimuth as the direction in which the strongest sound is directed, the simplest transfer function of each filter 306-309 is a real number between zero and one (including both limits).

Simple multiplication by a real number does not take into account the importance of pitch of sound with respect to directivity.
In a more versatile transfer function, the impulse is divided into predetermined frequency bands, and each frequency band is multiplied by a real amplification factor. A frequency band may be defined by a single number representing the highest frequency of the frequency band. Alternatively, certain real coefficients may be shown here for some frequency examples. This causes an appropriate interpolation to be applied between these frequencies (eg, frequency 400
Given an Hz and an amplification factor of 0.6 and a frequency of 1000 Hz and an amplification factor of 0.2, an amplification factor of 0.4 is obtained for a frequency of 700 Hz by direct interpolation).

In general, each filter 306-309 is a IIR or FIR filter (Infinite Impulse Responses) having a transfer function H represented by a Z-transform H (z).
e; Finite Impulse Response). Impulse X (t)
The following definition is obtained by the Z-transformation X (t) and the Z-transformation Y (t) of the impulse Y (t).

[0026]

(Equation 2)

Thus, it is sufficient to represent the coefficients [b ₀ b ₁ a ₁ b ₂ a ₂ ...] Used for modeling the Z-transform to represent any transfer function. The upper limits N and M used in the addition represent the precision required to define the transfer function. In practice, they store the coefficients used to model each single transfer function and / or
Or it depends on how much capacity is available for transmission in the transmission system.

FIG. 4 shows how the sound produced by the trumpet is directed. It is represented by the zero orientation and has eight frequency-dependent transfer functions and interpolation between them. The way in which sound is given directivity is three-dimensional, where the vertical axis represents volume in decibels, the first horizontal axis represents direction angles in degrees relative to zero azimuth, and the second horizontal axis represents sound frequency in kilohertz. Modeled in a coordinate system. For interpolation, the sound is represented by surface 400. At the top left end of the figure, the plane 400 is bounded by a horizontal line 401, which represents that the volume is independent of frequency in the zero azimuth direction. At the upper right end, surface 400 is bounded by a substantially horizontal line 402, indicating that the volume is independent of direction angle at very low frequencies (frequency near 0 Hz). The frequency response of the filter representing the various directional angles is a curve starting from line 402 and extending diagonally to the lower left of the figure. The direction angles are equidistant and their magnitudes are 22.5 °, 45 °, 67.5 °, 90 °, 112.
5 °, 135 °, 157.5 °, and 180 °. For example, curve 403
Represents the loudness as a function of frequency for sound propagating at an angle of 157.5 ° measured from zero azimuth, and this curve shows that in this direction the highest frequencies decay more than low frequencies.

According to the invention, the virtual acoustic environment is generated in a computer memory and processed with the same combination, or it is read from a storage medium such as a DVD disc (Digital Versatile Disc) and is used for audiovisual representation means ( It is suitable for playback on a local device for playback to a user via a display or a speaker. Further, the present invention can be applied to a system in which a virtual acoustic environment is generated by a device of a service provider and transmitted to a user via a transmission device. A device that reproduces directional sound processed by the method according to the present invention to a user and allows the user to select a point in the virtual acoustic environment where he wants to hear the reproduced sound is generally called a receiving device. This term is not limited to the present invention.

When the user has given the receiving device information about a point in the virtual acoustic environment where the user wants to hear the reproduced sound, the receiving device determines in which direction the sound is directed from the sound source to said point. In FIG. 4, when the receiving apparatus determines the angle between the zero direction of the sound source and the direction of the observation point as shown in the graph, the plane 400 is cut along a plane perpendicular to the frequency axis, and the direction angle axis is determined. Means to cut at that value, which is the angle between the zero azimuth and the observation point. The section between plane 400 and the perpendicular plane is a curve representing the relative loudness of sound detected in the direction of the observation point as a function of frequency. The receiving device forms a filter that realizes a frequency response based on the curve, and directs the sound generated by the sound source to the user through the filter formed before playing the sound to the user. If the user decides to change the position of the observation point, the receiving device determines a new curve and generates a new filter as described above.

FIG. 5 shows three virtual sound sources 501, 502 and 503 that are variously oriented.
5 shows a virtual acoustic environment 500 having Point 504 indicates the observation point selected by the user. To illustrate the situation shown in FIG. 5, in accordance with the present invention, a unique model representing how the sound is directed for each sound source 501, 502, and 503 is generated, whereby in each case, The model may be approximately as in FIGS. 3 and 4, but consider that the zero orientation has a different direction for each virtual source in the model. In this case, the receiving device needs to generate three separate filters to consider how the sound is directed. To generate the first filter, transfer functions that model how the sound transmitted by the first sound source is directed are determined, and with the aid of these transfer functions and interpolation, as in FIG. Is generated. Furthermore, the angle between the direction of the observation point and the zero orientation of the sound source 501 is determined, and with the help of this angle the frequency response in said direction on said surface can be read. The same operation is repeated separately for each sound source. The sound played to the user is the sum of the sounds from all three sources, in which each sound is filtered by a respective filter that models how the sound is directed.

According to the present invention, in addition to the actual sound source, sound reflection, in particular, early reflection can be modeled. In FIG. 5, a virtual sound source 506 formed by the virtual sound source method represents how the sound transmitted by the sound source 503 is reflected from a nearby wall.
This imaginary sound source can be processed in exactly the same way as the real sound source according to the invention, in other words, the directivity of the sound in the direction of the zero azimuth and in a direction different from the zero azimuth direction (frequency dependent if necessary) ) Can be determined. The receiver reproduces the sound generated by the imaginary sound source on the same principle as that used for the sound generated by the actual sound source.

FIG. 6 shows a system having a transmitting device 601 and a receiving device 602. The transmitting device 601 generates a virtual acoustic environment including at least one sound source and at least one spatial acoustic characteristic, and communicates the environment to the receiving device 602 in a certain format.
The transmission may be over a digital radio, television broadcast, or data network, for example. In transmission, the transmitting device 601 generates a recording such as a DVD (Digital Versatile Disc) based on the virtual sound environment that has already been generated, and the user of the receiving device obtains the recording at the time of use. Could mean. A typical application delivered as a recording is a concert by an orchestra whose sound source includes virtual musical instruments, a virtual or real concert hall where the space is modeled electrically, whereby the user of the receiving device with the device Can hear how the performance sounds in various places in the hall. If the virtual environment is audiovisual, it also includes a visual display implemented by computer graphics. In the present invention, the transmitting device and the receiving device do not need to be different devices, and the user generates a specific virtual acoustic environment with one device, and uses the same device to listen to what he generated himself. Can be.

In the embodiment shown in FIG. 6, a user of the transmitting device can use a virtual orchestra player with a computer graphics tool 603 and a corresponding tool 604 and a video animation, such as a musical instrument, with the help of a concert hall. Create a visual environment like In addition, he inputs via the keyboard 605 a transfer function that describes some directivity of the sound source of the environment he generated, and preferably how the sound is directed depending on the frequency. Modeling of how sound is directed can also be made based on measurements made on real sound sources. At that time, the directivity information is usually read from the database 606. The sound of the virtual instrument is loaded from the database 606. The transmitting device processes the information entered by the user, converts it into a bitstream in blocks 607, 608, 609, and 610 and combines the bitstream into one data stream in multiplexer 611. The data stream is provided to the receiving device 602 in a certain format. In the demultiplexer 612, the image section representing the still environment from the data stream is shown in a block 613, the time-dependent image section or animation is shown in a block 614, the time-dependent sound is shown in a block 615,
The coefficients representing the plane are then separated into blocks 616. The image sections are combined at display driver block 617 and provided to display 618. The signal representing the sound transmitted from the sound source is supplied from the block 615 to the filter bank 619. The filter bank 619 comprises a filter whose transfer function is reconstructed with the help of the parameters a and b obtained from the block 616. The sound generated by the filter bank is provided to headset 620.

FIGS. 7 a and 7 b show in more detail the configuration of a filter of a receiving device capable of realizing a virtual acoustic environment in a method according to the invention. In addition to the modeling of sound directivity according to the invention, other factors relating to sound processing are also taken into account in the figures. The delay unit 721 generates a mutual time difference between various sound components (for example, a sound reflected along various paths, or a mutual time difference between virtual sound sources located at various distances). At the same time, the delay means 721 operates as a demultiplexer that directs the correct sound to the correct filters 722, 723, and 724. Filters 722, 723, and 724 are filters characterized by the parameters described in more detail in FIG. 7b. The signals provided by them are, on the one hand, filters 701,
702 and 703, and on the other hand to adder 705 via adder and amplifier 704, which are echo branches 706, 707, 708 and 70
9, together with an adder 710 and amplifiers 711, 712, 713, and 714, so that post-echo can be generated for certain signals. The filters 701, 702, and 703 are, for example, HRTF models (Head-Relate
This is a directional filter that takes into account differences in the listener's sense of hearing in various directions based on d Transfer Function). Also, filters 701, 702, and 703
Most preferably includes a so-called ITD delay (Interaural Time Difference), which models the mutual time difference between sound components reaching the listener's ear from various directions.

In the filters 701, 702, and 703, each signal component is divided into left and right channels, and is generally divided into N channels in a multi-channel system. All signals associated with a channel are added to adders 715 or 716.
And is directed to summer 717 or 718, where the post-echo belonging to each signal is added to the signal. Lines 719 and 720 lead to a speaker or headset. In FIG. 7a, the points between filters 723 and 724 and between filters 702 and 703 mean that the present invention does not limit the number of filters in the filter bank of the receiver. There may be hundreds or thousands of filters depending on the complexity of the modeled virtual acoustic environment.

FIG. 7b shows in more detail the possibility of implementing a filter 722 characterized by the parameters shown in FIG. 7a. In FIG. 7b, the filter 722 includes three successive filter stages 730, 731 and 732, of which the first filter stage 730 represents propagation attenuation in a medium (usually air) and the second stage 731 is reflective. The third step 732 represents the absorption that occurs in the material, which is particularly applied when modeling reflections, and the sound propagates through the medium from the sound source to the observation point (possibly via the reflective surface) Consider both distance and characteristics of the medium, such as air humidity, pressure, and temperature. To calculate the distance, the first stage 730 obtains information about the position of the sound source in the coordinate system of the space to be modeled from the transmitting device and information about the coordinates of the point selected by the user as the observation point from the receiving device. . The first stage 730 obtains data representing the characteristics of the medium from either the transmitting device or the receiving device (the user of the receiving device can set the required media characteristics).
By default, the second stage 731 derives coefficients from the transmitting device that represent the absorption of the reflective surface, but in this case the user of the receiving device may be given the possibility to change the properties of the modeled space. The third stage 732 takes into account how the sound transmitted by the sound source is directed from the sound source in various directions in the modeled space. Thus, third stage 732 implements the invention presented in this patent application.

It has been generally described above how the properties of a virtual acoustic environment are processed and transmitted from one device to another by using parameters. Next, we discuss how the present invention applies to certain data transmission formats. Multimedia refers to the mutually synchronized presentation of audiovisual objects to a user. Conversational multimedia presentations are expected to become widespread in the future, for example, in the form of entertainment and electronic conferencing. There are a number of standards in the prior art that specify various methods of transmitting multimedia programs in electrical form. In this patent application, the so-called MPEG (Moti
on Picture Experts Group). The MPEG-4 standard, which is being developed at the time this patent application is filed, of that standard has the goal that transmitted multimedia presentations can include real or virtual objects that together form an audiovisual environment. . The invention is by no means limited to use in connection with the MPEG-4 standard, but may be applied, for example, to extensions of the VRML97 standard or even to future audiovisual standards that are currently unknown.

A data stream based on the MPEG-4 standard can include multiplexed sections in which time (such as a synthetic sound) and parameters (such as the position of a sound source in the space to be modeled) are continuous. Includes audiovisual objects. Since objects can be defined to be hierarchical, so-called primitives are at the lowest level of the hierarchy. In addition to the objects, multimedia programs based on the MPEG-4 standard contain so-called scene descriptions, which contain information about the interrelationship of the objects and information about the arrangement of the general settings of the program, and are very useful. That information is encoded and decoded separately from the actual object. The scene description is a BIFS section (binary format for the scene description)
It is called. The transmission of the virtual acoustic environment according to the present invention is based on the MPEG-4 standard (SAOL /
SASL: Structured Audio Orchestra Language / Structured Audio Score Langu
age) or a structured speech language defined in the VRML97 language.

In the above-mentioned languages, sound nodes that model sound sources are currently defined. According to the invention it is possible to define an extension of a known sound node, which is referred to in the present patent application as a DirectiveSound node. In addition to the known sound nodes, it further includes a field, referred to as a directional field, which supplies information necessary to reconstruct a filter representing the directivity of the sound. Three different alternatives for modeling filters have been described above. The following describes how these alternatives are implemented in the directivity field of the pointing sound node according to the present invention.

According to a first alternative, each filter modeling in a direction different from a certain zero azimuth corresponds to a simple multiplication by an amplification factor which is a normalized real number between 0 and 1.
At that time, the contents of the directivity field are as follows, for example. ((0.79 0.8) (1.57 0.6) (2.36 0.4) (3.140
. 2))

In this alternative, the directional field includes the same number of value pairs as directions that are different from the zero orientation in the sound source model. The first number in a pair of numbers indicates the angle in radians between the direction of interest and the zero azimuth, and the second number indicates the amplification factor in said direction.

According to a second alternative, the sound in each direction different from the zero azimuth direction is divided into frequency bands, each of which has its own amplification factor. The contents of the directional field are
For example: ((0.79 125.0 0.8 1000.0 0.6 4000.0 0.4)
(1.57 125.0 0.7 1000.0 0.5 4000.0 0.3) (2.36 125.0 0.6 1000.0 0.4 4000.0 0.2) (3.14) 125.0 0.5 1000.0 0.3 4000.0 0.1))

In this alternative, the directional field comprises a set of values separated from each other by the same number of inner braces as directions that are different from the zero orientation in the sound source model.
In each set of numbers, the first number indicates the angle in radians between the direction of interest and the zero azimuth. Following the first number are pairs of numbers, the first of which indicates a certain frequency in hertz and the second is the amplification factor. For example, a set of numerical values (0.79 125.0 0.8 1000.0 0.6 4000.0
0.4) means that an amplification factor of 0.8 in the direction of 0.79 radian has a frequency of 0 to 12
It can be interpreted that an amplification factor of 0.6 is used for frequencies 125-1000 Hz, and an amplification factor of 0.4 is used for frequencies 1000-4000 Hz. Alternatively, the above set of numbers may have a gain of 0.8 at a frequency of 125 Hz and a gain of 1000 Hz in the 0.79 radian direction.
Use notation that means that the gain is 0.4 at a frequency of 4000 Hz and the gain at other frequencies is calculated from these by interpolation and extrapolation. Is possible. With respect to the present invention, it is not essential which notation is used, as long as the notation used is known to both the transmitting device and the receiving device.

According to a third alternative, the transfer function is applied in each direction different from the zero azimuth and given its coefficients a and b of the Z-transform to define the transfer function. The contents of the directivity field are, for example, as follows. _{((45 b 45.0 b 45.1 a} 45.1 b 45.2 a 45.2 ...) (90 b 90.0 b 90.1 a 90.1 b 90.2 a 90.2 ...) (135 b 135.0 b 135.1 a 135.1 b 135.2 a 135.2 ...) (180 b 180.0 b 180.1 a _180.1 b _180.2 a _180.2 …))

In this alternative, the directivity field also includes a set of values separated from each other by the same number of inner braces as a plurality of directions different from the direction of the zero bearing in the sound source model. In each set of numbers, the first number indicates the angle between the direction of interest and the zero azimuth, in degrees this time. In this case, other known angular units can be used as well, as in the case described above. After the first number are the coefficients a and b which determine the Z-transform of the transfer function used in the direction of interest. The point after each set of values is that the invention defines the coefficients a and b that define the Z-transform of the transfer function.
Means that no limit is imposed on the number of. In each different set of values, there can be different numbers of coefficients a and b. In a third alternative, the coefficients a and b may also be given as their own vectors. To that end, efficient modeling of FIR or all-pole IIR filters is described in the publication Ellis, S.A. (Ellis, S.
1998: "Towards more realistic sounds in VMRL (Towards more realistic)
sound in VMRL), Proc. VRML'98, Money, USA, February 16-19, 1998, pp. 95-100.

The embodiments of the invention presented above are, of course, intended only as examples, and they have no effect on limiting the invention. In particular, the way in which the parameters representing the filters are arranged in the directional field of the DirectiveSound node can be chosen in a very large number of ways.

[Brief description of the drawings]

FIG. 1 is a diagram showing a known directional acoustic model.

FIG. 2 is a diagram showing another known directional acoustic model.

FIG. 3 is a diagram schematically showing a directional acoustic model according to the present invention.

FIG. 4 is a graph showing in which direction a sound generated by a model according to the present invention is directed.

FIG. 5 is a diagram showing how the present invention is applied to a virtual acoustic environment.

FIG. 6 shows a system according to the invention.

FIG. 7a shows a part of the system according to the invention in more detail.

FIG. 7b shows a detail of FIG. 7a.

──────────────────────────────────────────────────続 き Continuation of front page (81) Designated country EP (AT, BE, CH, CY, DE, DK, ES, FI, FR, GB, GR, IE, IT, LU, MC, NL, PT, SE ), OA (BF, BJ, CF, CG, CI, CM, GA, GN, GW, ML, MR, NE, SN, TD, TG), AP (GH, GM, KE, LS, MW, SD, SL, SZ, UG, ZW), EA (AM, AZ, BY, KG, KZ, MD, RU, TJ, TM), AE, AL, AM, AT, AU, AZ, BA, BB, BG, BR , BY, CA, CH, CN, CU, CZ, DE, DK, EE, ES, FI, GB, GD, GE, GH, GM, HR, HU, ID, IL, IN, IS , JP, KE, KG, KP, KR, KZ, LC, LK, LR, LS, LT, LU, LV, MD, MG, MK, MN, MW, MX, NO, NZ, PL, PT, RO, RU, SD, SE, SG, SI, SK, SL, TJ, TM, TR, TT, UA, UG, US, UZ, VN, YU, ZA, ZWF terms (reference) 5D062 CC12 CC16

Claims (14)

[Claims] 1. A method for processing a virtual acoustic environment in an electronic device, whereby the virtual acoustic environment comprises at least one sound source (300), in which direction the sound is directed A set of filters (
306, 307, 308, 309) are provided with a set of direction-dependent filters corresponding to the sound sources so that the effect on sound depends on predetermined parameters of each filter. how to. 2. A reference direction (301) and a different set of directions (302).
, 303, 304, 305) are defined for the sound source, whereby the filters are determined such that the effect on the sound of the set of filters (306, 307, 308, 309) depends on the parameters for each filter. The method according to claim 1, wherein the method is provided corresponding to each direction different from the reference direction. 3. The method of claim 2, wherein the parameter for each filter is a set of amplification factors for determining a respective amplification of sound directed from the sound source in various directions. 4. The method of claim 3, wherein the set of amplification factors includes different amplification factors for various frequencies of sound in at least one determined direction different from the reference direction. 5. The method according to claim 2, wherein the parameter for each filter is a coefficient [b ₀ b ₁ a ₁ b ₂ a ₂ ...] Of the following fractional expression representing the Z-transform of the transfer function of the filter. . (Equation 1) 6. A method according to claim 1, further comprising the step of: modeling the manner in which the sound is directed in other directions than the reference direction and in each of the determined directions different from the reference direction. 3. The method of claim 2, including interpolating (400) between each provided filter. 7. The transmitting device generates certain virtual acoustic environments (500) that include sound sources (501, 502, 503, 504), so that the manner in which sound is directed from these sources in a certain direction affects the sound. Wherein the influence is modeled by filters dependent on the parameters for each filter; the transmitting device transmits information about the parameters for each filter to a receiving device; and Generating a filter bank including filters whose influence on the acoustic signal depends on parameters related to each filter, and generating parameters related to each filter based on information transmitted by the transmitting device. The described method. 8. The method according to claim 7, wherein the transmitting device transmits information about the parameters for each filter to the receiving device as part of a data stream based on the MPEG-4 standard. 9. The method according to claim 1, wherein the sound source is an actual sound source (501, 502, 503). 10. The method of claim 1, wherein the sound source is a reflection (504). 11. A system for processing a virtual acoustic environment including at least one sound source, characterized by parameters to model in which direction a sound is directed from a sound source belonging to the virtual acoustic environment. A means for generating a filter bank (619) containing the filters to be obtained. 12. The system according to claim 11, comprising a transmitting device (601), a receiving device (602), and means for realizing electrical communication between the transmitting device and the receiving device. 13. A multiplexing means (611) for adding a parameter representing a filter characterized by a parameter in the transmitting device to a data stream based on the MPEG-4 standard, and a filter characterized by a parameter in the receiving device. Demultiplexing means (612) for detecting a parameter representing the following from a data stream based on the MPEG-4 standard.
The described method. 14. A multiplexing means (611) for adding a parameter representing a filter characterized by a parameter in the transmitting device to a data stream based on the extended VRML97 standard, and representing a filter characterized by a parameter in the receiving device. 12. Demultiplexing means (612) for detecting parameters from a data stream according to the extended VRML97 standard.
The described method.